Research map / Vision-language and multimodal / Robot action and navigation
ImageBrush: Learning Visual In-Context Instructions for Exemplar-Based Image Manipulation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add diffusion-based progressive denoising instead of masked image modeling (ablation) | ImageBrush · image manipulation | prompt fidelity FID (lower better): 78.65 -> 23.97 | Helped |
| add diffusion-based progressive denoising instead of masked image modeling (ablation) | ImageBrush · image manipulation | image fidelity FID (lower better): 82.33 -> 24.43 | Helped |
| add visual prompt cross-attention feature injection (ablation) | ImageBrush · image manipulation | prompt fidelity FID (lower better): 39.44 -> 23.97 | Helped |
| add visual prompt cross-attention feature injection (ablation) | ImageBrush · image manipulation | image fidelity FID (lower better): 41.51 -> 24.43 | Helped |
| add region-of-interest bounding box input (ablation) | ImageBrush · image manipulation | prompt fidelity FID (lower better): 24.29 -> 23.97 | Helped |
| add region-of-interest bounding box input (ablation) | ImageBrush · image manipulation | image fidelity FID (lower better): 25.74 -> 24.43 | Helped |
| increase number of in-context examples from 1 to 3 (ablation) | ImageBrush · image manipulation | prompt and image fidelity | Helped |
| shuffle order of in-context examples (ablation) | ImageBrush · image manipulation | prompt and image fidelity | No clear effect |
Closest papers by meaning
- Unleashing In-context Learning of Autoregressive Models for Few-shot Image Manipulation (CVPR 2025 · similarity 0.90)
- Guiding Instruction-based Image Editing via Multimodal Large Language Models (ICLR 2024 · similarity 0.89)
- Instruction-based Image Manipulation by Watching How Things Move (CVPR 2025 · similarity 0.89)
- Text-Vision Co-Instructed Image Editing (NeurIPS 2026 · similarity 0.89)
- PERIA: Perceive, Reason, Imagine, Act via Holistic Language and Vision Planning for Manipulation (NeurIPS 2024 · similarity 0.89)
- Learning to Manipulate Artistic Images (AAAI 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.